Artificial Intelligence Expert System Based on Continuous Glucose Monitoring (CGM) Data for Auto-Adaptive Adjustment Therapy Protocol – How to Make Sensors and Patients to Think Forward and Work Together?

Worldwide adoption of continuous glucose monitoring (CGM) portable miniature sensors in routine clinical practice is increasing for type 1 diabetes treatment in children. The commonly used sensors provide the blood glucose concentration and its change rate in real time (samples every 5 minutes or even 1 minute). Due to these sensors we are in position to collect a significant amount of data from various patients. To better process the data, artificial intelligence (AI) techniques are used to personalize the patient's recommendation and to create decision support expert system with self-adjustment. The focus of this paper is the system’s architecture and use case scenarios of a research project aimed to develop an Artificial Intelligence expert system based on continuous glucose monitoring (CGM) data. The project is developed by a mixed team composed by medical doctors in France and computer science specialists in Romania. By regularly analyze and interpret the CGM data, the decision support expert system will generate detailed reports for doctor and patient use and real-time alerts. One of the key components is the implementation of an auto-adaptive adjustment therapy protocol for type 1 diabetes treatment in children.